effect of a core conditioning program on lumbar paraspinal
TRANSCRIPT
University of Tennessee, Knoxville University of Tennessee, Knoxville
TRACE: Tennessee Research and Creative TRACE: Tennessee Research and Creative
Exchange Exchange
Masters Theses Graduate School
12-2014
Effect of a Core Conditioning Program on Lumbar Paraspinal Effect of a Core Conditioning Program on Lumbar Paraspinal
Area, Asymmetry and Pain Score in Military Working Dogs with Area, Asymmetry and Pain Score in Military Working Dogs with
Lumbosacral Pain Lumbosacral Pain
Andrea Leigh Henderson University of Tennessee - Knoxville, [email protected]
Follow this and additional works at: https://trace.tennessee.edu/utk_gradthes
Part of the Small or Companion Animal Medicine Commons
Recommended Citation Recommended Citation Henderson, Andrea Leigh, "Effect of a Core Conditioning Program on Lumbar Paraspinal Area, Asymmetry and Pain Score in Military Working Dogs with Lumbosacral Pain. " Master's Thesis, University of Tennessee, 2014. https://trace.tennessee.edu/utk_gradthes/3155
This Thesis is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected].
To the Graduate Council:
I am submitting herewith a thesis written by Andrea Leigh Henderson entitled "Effect of a Core
Conditioning Program on Lumbar Paraspinal Area, Asymmetry and Pain Score in Military
Working Dogs with Lumbosacral Pain." I have examined the final electronic copy of this thesis
for form and content and recommend that it be accepted in partial fulfillment of the
requirements for the degree of Master of Science, with a major in Comparative and
Experimental Medicine.
Darryl L. Millis, Major Professor
We have read this thesis and recommend its acceptance:
Silke Hecht, Marti S. Drum
Accepted for the Council:
Carolyn R. Hodges
Vice Provost and Dean of the Graduate School
(Original signatures are on file with official student records.)
Effect of a Core Conditioning Program on Lumbar Paraspinal Area, Asymmetry and Pain Score
in Military Working Dogs with Lumbosacral Pain
A Thesis Presented for the
Master of Science
Degree
The University of Tennessee, Knoxville
Andrea Leigh Henderson
December 2014
ii
Acknowledgements
I would like to express my sincere appreciation to Dr. Darryl Millis, Dr. Silke Hecht, Dr. Marti
Drum and Dr. Irfan Asif for taking part in my committee for this project, and for all their input
into experimental design, data analysis and writing. In particular, Dr. Hecht spent a great deal of
time measuring muscle cross-sectional area and density, going above and beyond her regular
clinical duties. Dr. Millis traveled to San Antonio with me in order to facilitate smooth
execution of the initial orthopedic and neurologic examinations.
I would also like to thank Ann Reed for her statistical consultation, assistance with developing
models and answering my numerous questions pertaining to the testing design.
Dr. Jeryl Jones played an instrumental role in this project with her experience in lumbar
paraspinal muscle evaluation and measurement. She provided me with her expert guidance and
her protocol for collecting quality CT scans and measurements.
Finally, I would like to acknowledge the work of several veterinary officers, technicians and
handlers for their time and dedication outside of numerous regular duties. MAJ Sean Majoy,
MAJ Patrick Grimm, Ms. Robyn Miller, Ms. Kelley Meyer and PFC Kristi Deal were especially
involved in conducting the experimental design. I could not possibly have completed this project
from afar without the tremendous efforts of those involved at the DoD Military Working Dog
Center at Lackland Air Force Base.
iii
Abstract
Introduction: Lumbosacral pain and stenosis are common causes of retirement from duty for
Military Working Dogs (MWDs). Working dogs that receive surgical management for this
condition often have a poor prognosis for return to duty after recovery. Humans with chronic
low back pain demonstrate paraspinal muscle asymmetry, pain and dysfunction that often
improve in response to an exercise program. This study investigated whether dogs with mild
lumbosacral pain have decreased lumbar paraspinal muscle area, symmetry, and density, as well
as increased pain and dysfunction compared to control dogs. Additionally, response of pain and
dysfunction to an exercise program was assessed.
Materials and Methods: Visual Analog Scale (VAS) scores for lumbosacral pain, functional
questionnaire scores for search and detection tasks, and computed tomography images were
evaluated for eight MWDs with lumbosacral pain along with eight control dogs. Mean cross-
sectional muscle area (CSA)-to-vertebral ratio, asymmetry and density were determined for five
lumbar paraspinal muscles bilaterally at the L5, L6 and L7 caudal endplates. Four dogs with
lumbosacral pain rested and four dogs completed an eight-week core stabilizing exercise
program. Repeated assessments of lumbosacral pain, dysfunction and muscle parameters for
dogs with lumbosacral pain were made at the conclusion of the exercise program.
Results: The multifidus lumborum and longissimus lumborum muscles demonstrated
significantly reduced CSA (p= 0.020, p = 0.021, respectively) in dogs with lumbosacral pain.
Muscle density was decreased in dogs with lumbosacral pain for multifidus lumborum (p =
0.030) and quadratus lumborum (p = 0.011). Multifidus lumborum muscle CSA (p = 0.019),
symmetry (p = 0.002) and density (p = 0.024) were significantly higher than at baseline for dogs
with LS pain after completion of the exercise program. Functional questionnaire scores improved
significantly for exercised dogs (p = 0.031) but did not improve for rested dogs (p = 0.828).
Discussion: Military Working Dogs with mild lumbosacral pain and dysfunction had
significantly smaller CSA, symmetry and density for both multifidus lumborum and longissimus
lumborum muscle groups. An 8-week core strengthening program was associated with
significantly improved performance in evaluated tasks for dogs with lumbosacral pain.
iv
Table of Contents
Chapter 1: Introduction…………………………………………………………………………...1
Chapter 2: Methods and Materials………………………………………………………………..7
Chapter 3: Results……………………………………………………………………………….17
Chapter 4: Discussion…………………………………………………………………………...27
References………………………………………………………………………………………..36
Appendices………………………………………………………………………………….……44
Appendix A: Orthopedic and Neurologic Examination………………………….….…..45
Appendix B: Functional Assessment Questionnaire…………………………………… 47
Appendix C: Exercise Protocol………………………………………………………….52
Vita……………………………………………………………………………………………….56
v
List of Figures
Figure 1: Goniometer Positioning for Evaluation of Tail Hyperextension……...…………....….8
Figure 2: Positioning of Military Working Dogs for CT of the Lumbar Spine……………….…11
Figure 3: Sample Image Slice at the Caudal Endplate of L5…………………………………….12
Figure 4: Goniometric Angle of Extension and Pressure at L7-S1.……………………….……18
Figure 5: Correlation between VAS and Functional Questionnaire Score……………….……..19
Figure 6: Baseline Muscle:Vertebral CSA Ratio for LS Pain and Control Dogs………….……21
Figure 7: Baseline Muscle ASI for LS Pain and Control Dogs…………………………….…...21
Figure 8: Baseline Muscle Density for LS Pain and Control Dogs……………………….…….22
Figure 9: Functional Questionnaire and VAS Scores Before and After Exercise……….….…..23
Figure 10: Multifidus and Longissimus CSA, ASI and Density Before and After Exercise…...24
1
Chapter 1: Introduction
Degenerative lumbosacral stenosis (DLSS) is the most common pathologic condition of the
lumbar spine in large-breed dogs. It is one of several possible etiologies of cauda equina
syndrome, a degenerative condition that results in compression of the nerve roots composing the
lumbosacral trunk. Intervertebral disc degeneration at L7-S1 plays a large role in the
pathogenesis of DLSS. Congenital or activity-related hypermobility at this joint may facilitate
L7-S1 disc degeneration, contributing to instability. Surrounding soft tissues proliferate in an
attempt to stabilize the region, worsening spinal canal stenosis, neuropathic pain, and further
alterations in biomechanics. A continuous cycle of pain and neurologic dysfunction occurs due
to compression and compromised blood supply to the nerve roots (Meij & Bergknut 2010).
German Shepherd Dogs and structurally related breeds, such as the Belgian Malinois, appear to
be particularly susceptible to DLSS; Shepherds represent an estimated 25–57% of all breeds
evaluated for DLSS (Daniellson & Sjostrum 1999, De Risio et al. 2001, Suwankong et al. 2008).
The reason for this breed predisposition is not well-understood, but in one study, German
Shepherd Dogs demonstrated reduced spinal mobility at L7-S1, which was attributed to reduced
angulation of articular facets (Benninger et al. 2006). Such structural alterations may contribute
to disc degeneration in this breed. Males are over-represented among dogs with DLSS, with
odds ratios from 1.3:1 to 5:1 reported. Clinical signs of DLSS most often develop in middle-
aged or older dogs, with an average presenting age of 7 years. Clinical history may include
pelvic limb lameness and pain or difficulty on rising from recumbency, climbing stairs, or
jumping; tail hypotonia or urinary or fecal incontinence (Meij & Bergknut 2010).
2
Spinal pain has been suggested to be the third most common reason for working dog retirement
(Moore et al. 2001). Specifically, DLSS and lumbosacral pain are common in the Department of
Defense (DoD) military working dog population and are a frequent cause of retirement of dogs
from active duty. A number of surgical procedures have been performed on military working
dogs to correct lumbosacral instability and/or disc extrusion that may be the source of pain in this
area. However, Linn et al. (2003) found that only 41% of 29 American military working dogs
with DLSS successfully returned to regular duty after surgical management. The prognosis for
return to duty was negatively correlated with age and severity of signs. In many cases, dogs are
not surgical candidates due to age, severity of the condition, logistics and expected poor
continued working life without surgery.
Degenerative lumbosacral stenosis (DLSS) in dogs has similar manifestations of pain and
neurologic dysfunction to those attributed to chronic low back pain (CLBP) in humans. Despite
obvious differences in posture between bipeds and quadrupeds, humans and dogs share several
spinal biomechanical characteristics, including similarities in axial compressive loads (Smit
2002, Zimmerman et al. 1992) and in the pathogenesis of disc degeneration in non-
chondrodystrophic breeds (Benninger et al. 2006). CLBP is one of the leading causes of
disability in working people, and is associated with muscle atrophy and asymmetry, pain, and
dysfunction (Gibbons et al. 1997, Kamaz et al. 2007, Parkkola et al.1992, Danneels et al. 2000,
Kader et al. 2000, Hides et al. 2008, Marshall et al. 2011, van Dieen et al. 2003). Frequently
clinical signs of CLBP are not well-correlated with specific abnormalities, such as lumbar disc
degeneration, on advanced imaging (Beattie et al. 2000, Takatalo et al. 2011), and are treated
non-surgically. Several studies have shown that symptoms of chronic low back pain in people
3
frequently respond to paraspinal muscle strengthening programs (Hides et al. 2008, Marshall et
al. 2006), though the treatment response may vary depending on the mechanism of the pain.
Danneels et al. (2000) postulated that pain-guarding behavior, reflex inhibition, or inflammation
may result in reduced activation of paraspinal muscles, leading to disuse atrophy. Renkawitz
(2006) identified altered paraspinal neuromuscular activation patterns in athletes with low back
pain.
Several muscle groups are associated with the lumbar spine and are thought to contribute to
spinal stability. These are generally categorized as either deep (local) or superficial (global)
muscles, based on their anatomic location relative to the vertebral axis. Deep muscles such as
the lumbar multifidus contribute to spinal stabilization and may be dysfunctional in human
patients with CLBP. However, global muscles are thought to provide compensatory stabilization
in the face of dysfunctional deep muscle groups, leading to abnormal activation patterns (Barr et
al. 2005).
Several studies have identified selective, significant atrophy of the lumbar multifidus muscle in
human CLBP patients (Danneels et al. 2000, Kamaz et al. 2007), even in elite athletes (Hides et
al. 2008). The lumbar multifidus is thought to be the most important muscle for lumbar
segmental stability in humans because it is the largest paraspinal muscle in the region with the
most medial location, flanking the dorsal spinous processes. The multifidus contributes the
most of all the muscle groups to stability and control of neutral zone movement in the spine, and
in one study, magnetic resonance imaging (MRI) showed that lumbar multifidus muscle mass
was atrophied in 80% of patients with disc degeneration and nerve compression (Kader et al.
4
2000). Kamaz et al. (2007) found that cross-sectional area (CSA) of the multifidus, psoas, and
quadratus lumborum was significantly lower in patients with CLBP than in controls. Atrophy
was most prominent in the lumbar multifidus, which agrees with other findings (Bouche et al.
2011, Danneels et al. 2000). Another small study (Gibbons et al. 1997) found no significant
difference in muscle CSA between CLBP patients and healthy controls, but found degenerative
changes within the muscles. Atrophy of the psoas major muscle has been demonstrated as well,
but has not always coincided with the symptomatic side of the body in the case of lateralized
signs. Nonetheless, shifting, bilateral pain frequently occurs in association with CLBP in
humans (Kamaz et al. 2007).
The quadratus lumborum muscle acts synergistically on lumbar vertebrae with the psoas and
deep erector spinae muscles, facilitating lumbar and pelvic stability in all major planes of
motion. The psoas major keeps the human body erect in all three planes, and the gluteus
maximus contributes to spinal stability through the thoracolumbar fascia (Kamaz et al. 2007). In
dogs, it has been suggested that the quadratus lumborum is responsible for stabilization of the
lumbar portion of the vertebral column by restricting lateral spinal flexion (Hermanson & Evans
1993).
Lumbar stabilization programs for treatment of CLBP must be multifaceted to address the
associated muscle groups and dysfunctions. Deeper, local muscle groups are often the target for
therapeutic interventions for CLBP, but likely global muscle stabilization is necessary as well.
Appropriate lumbar muscular function has been shown to overcome structural abnormalities in
the spinal column (Hides et al. 2008, Panjabi 1992, Barr et al. 2005). Additionally, lumbar
5
pelvic stabilization is closely related to global body awareness, proprioception and balance
(Hodges et al. 2004), and patients with CLBP have been shown to have reduced performance on
balance tests (Ebenbichler et al. 2001). Therefore, improving these components should be
among the goals of a conditioning program to address CLBP. Development of consistently
appropriate interventions requires a thorough understanding of the roles of the paraspinal trunk
muscles in normal motion and in the pathogenesis of CLBP.
Similarly, characterization of canine DLSS-affected muscle groups may be useful for
determining ways to redistribute axial and shear loads imposed upon the lumbar spine and
improve pain and function. Therefore, dogs that are mildly affected by DLSS may benefit from
paraspinal muscle strengthening if asymmetry and reduced muscle CSA reflect abnormal
activation patterns that contribute to the clinical manifestations of pain and dysfunction.
Core stabilizing and strengthening exercise programs have been shown to increase lumbar
paraspinal muscle mass/symmetry, as well as improve pain and return to function in people with
CLBP (Kim et al. 2011, Marshall et al. 2006, Niemisto et al. 2003, O’Sullivan et al. 1998).
Such activities are designed not to treat the underlying source of the pain, but to promote spinal
stability, muscular control, and reverse pathological paraspinal muscle activation patterns that
contribute to prolonged pain and reduced function. Both anticipatory and responsive muscle
activation patterns are targeted in traditional programs (Carneiro et al. 2010). Such exercises
include combinations of dynamic and sustained contractions of deep abdominal and local and
global lumbar paraspinal muscles. These contractions can be achieved through postures and
exercises for which balance is required in the face of instability. Additional exercises are
6
designed to promote extension, lateral flexion and ventroflexion of the spine. Core stabilization
activities for dogs may include balancing on such surfaces as physioballs, balance boards and
inflated disks, walking up and down stairs or ramps, circling, weaving or performing spinal
movements to follow a reward while standing on an unstable surface.
There are currently no published studies on the effects of core stabilization exercises on epaxial
musculature and spinal pain in dogs. A small pilot study has demonstrated increase in lumbar
multifidus muscle CSA of three healthy dogs after an 8-week program of stabilization exercises
on a physioball. The muscle CSA was estimated using ultrasound, and there was an insufficient
number of cases to achieve significance (Teeling and Van den Berg 2012). One equine study
found that dynamic spinal mobilization exercises in eight horses without neck or back pain
increased the thoracolumbar multifidus paraspinal muscle cross-sectional area and symmetry,
though no control group was used (Stubbs et al. 2011).
The objective of the study reported here was to determine whether lumbar paraspinal muscle
cross-sectional area (CSA), symmetry and density are decreased in military working dogs with
lumbosacral pain as compared with normal dogs. Additionally, the study evaluated effects of an
eight-week core conditioning program on muscle mass, symmetry, pain and function in dogs
with DLSS.
7
Chapter 2: Materials and Methods
Subject Assessment
Records of 114 military working dogs with no history of orthopedic or neurologic disease were
evaluated from within the resident population of canine training aids available at the Department
of Defense (DoD) Military Working Dog Veterinary Services. Participants in the study were
required to be one of the following three breeds: German Shepherd Dog, Belgian Malinois or
Dutch Shepherd Dog. Additional inclusion criteria for selected dogs were a normal orthopedic
and neurologic examination, age between 5 and 11 years, single-purpose training in explosives
detection, a temperament that would allow handling during the exercise program, and absence of
administration of any analgesic or anti-inflammatory medications for eight weeks prior to the
beginning of the study.
A complete physical, orthopedic and neurologic examination was performed for each dog by a
single veterinarian (A.H.). The evaluation included subjective gait analysis and assessment of
lumbosacral pain based on digital pressure, tail hyperextension and lumbar hyperextension.
Additionally, objective data collected on examined dogs included goniometric tail elevation
angle and discomfort generated by dorso-ventral pressure over L7-S1 using an algometer (Pain
Diagnostics and Thermography, Great Neck, New York, USA). For the goniometric tail
elevation angle, zero degrees was represented by a line drawn from the tuber sacrale to the tuber
ischium, with the axis of rotation positioned at the midline of the tail base (Figure 1). Dogs were
excluded from the study if any of the following abnormalities were found: visible lameness
8
associated with any physical exam findings not consistent with LS pain, paresis or ataxia,
discomfort, abnormal range of motion or abnormal palpation findings on the orthopedic exam
(Appendix A). Additionally, dogs were eliminated if they demonstrated any of the following
upon neurological exam (Appendix A): proprioceptive deficits, abnormal myotatic, flexor or
perineal reflexes, or abnormal postural reactions.
After the initial screening orthopedic and neurologic evaluation for selection of cases, dogs
returned to their normal daily activities for 10 weeks before performing the baseline CT scans
and pain and function assessments. Dogs were removed from the study if they underwent any
change in activity, analgesic, anti-inflammatory or chondroprotective agent administration
during the study. Dogs were also removed from the study if they demonstrated illness, obvious
discomfort or other adverse effects deemed sufficient by the attending veterinarian to warrant
rescue analgesia, rest or surgical intervention. This study received approval by the DoD Military
Figure 1: Illustration showing goniometer position with fixed arm bisecting the wing
of the ilium and axis of rotation at the base of the tail. A smaller angle (degrees)
represents more tail hyperextension.
9
Working Dog Veterinary Services Institutional Animal Care and Use Committee prior to
initiating the experimental design.
Outcome Measures
A single blinded observer (S.M.) completed a baseline functional assessment and visual analog
scale (VAS) assessment for lumbosacral pain in all dogs after the ten-week period of baseline
activity, followed by radiographic and CT imaging. The following VAS scoring guidelines were
provided to the assessor:
1. The VAS was defined using the number of millimeters past zero on a continuous
100mm line with “no pain” marked at zero and “maximum possible pain” marked at
100.
2. VAS score was based on subjective evaluation of physical exam findings, including:
a. pain on lordosis test (hyperextension) of the lumbar spine with hips in flexion
b. pain upon mildly or moderately applied ventral digital pressure over the L7-
S1 disc space or articular facets
c. subjective lameness identified in one or both pelvic limbs that cannot be
attributed to any orthopedic findings on physical exam at a walk or trot on a
flat horizontal surface or during circling in clockwise and counterclockwise
directions at a walk and trot. Circles had a diameter of approximately 6-8 feet
around the handler
d. pain on dorsal tail base elevation
Following each VAS scoring assessment, the same evaluator observed each dog performing
normal tasks required for search and detection training or operations. Assessment was captured
10
quantitatively using a 10-item functional assessment questionnaire designed specifically for this
study to evaluate military working dogs during tasks required for detection (Appendix B, MWD
Functional Questionnaire). Activities assessed included jumping into a position in which
forelimbs were elevated with feet at the height of the withers, jumping onto and off of an
obstacle at the approximate height of a vehicle interior, sit-to-stand and sit-to-down, and
navigation of obstacles including a 2-foot jump, a double stairway (8 x 23 feet) and a narrow
dog-walk 18 feet in length. Each dog was encouraged to perform five trials of each activity. If
the assessor observed significant discomfort or inability for the dog to perform the task, the task
ended without completion of five trials and the assessor completed the question based on the
number of trials that had been attempted up to that point. All military working dogs had past
experience with the obstacle course because the obstacles are included in their normal training
protocols. Both outcome measures (VAS scores and functional questionnaires) were used to
assign dogs to each study group (LS pain positive or LS pain negative), as well as to evaluate
progress in dogs with LS pain after the 8-week exercise program. Dogs with a VAS score of less
than or equal to 10% and a functional questionnaire score of less than or equal to four were
placed in the control study group; dogs with values greater than these were considered “LS pain
positive.”
Radiographic and Computed Tomography (CT) Imaging
All dogs receiving imaging were sedated with dexmedetomidine HCl (DexDomitor®, Zoetis, NJ,
USA) and butorphanol at 0.003-0.007 mg/kg and 0.3mg/kg, respectively. Dogs were
11
administered atipamezole (Antisedan®, Zoetis, NJ, USA) at a volume equal to that of the
dexmedetomidine following completion of the imaging studies.
Lateral and ventrodorsal pelvic limb radiographs were made under sedation prior to each CT
scan. Dogs were excluded from the study if pelvic films demonstrated radiographic evidence of
hip osteoarthritis beyond subtle osteophytes, transitional vertebrae or other musculoskeletal
abnormalities expected to overlap with the clinical signs of degenerative lumbosacral stenosis.
Figure 2: Positioning of military working dogs for CT of the lumbar spine in A, flexion
of the hips at 50 +/- 3 degrees and B, extension of the hips to 145 +/- 3 degrees
12
All dogs were placed in dorsal recumbency for the CT scans (Figure 2). CT scans were
performed using a 64-slice volume CT scanner (GE Light-Speed VCT-XT; GE Healthcare,
U.K.) at 120 kV and 50 mA with 1.25 mm slice thickness, immediately following pelvic limb
radiographs for all dogs selected to remain in the study. Two separate lumbar CT scans were
acquired from L1-L2 to the level of the pelvic ischium, the first with the pelvic limbs at 145
degrees (+/- 3 degrees) of hip extension and the second with the pelvic limbs at 50 degrees (+/- 3
degrees) of hip flexion. All CT scans were collected by a board-certified veterinary radiologist
(P.G.). A density calibration phantom (Image Analysis QCT-Bone Mineral Phantom; Image
Analysis, KY, USA) was included in the field of view for each scan as part of routine research
CT scan protocols at the DoD Military Working Dog Center.
Figure 3: Sample CT image slice at the caudal endplate of L5. Cross-sectional areas of
the vertebral body (L5) and left and right multifidus lumborum (MF), longissimus
lumborum (LL), quadratus lumborum (QL) and iliopsoas (IP) were measured.
MF LL
IP QL
L5
13
A commercial picture archival and communication system (SECTRA PACS IDS7; Sectra
Medical Systems AB, Sweden) was used for image viewing and measurements. CT information
was evaluated and measured by two individuals who were blinded to the treatment group
assignment of each dog (S.H. and A.H). Cross-sectional area (mm2) and density in Hounsfield
Units (HU) were measured from transverse sections of the left and right multifidus lumborum,
longissimus lumborum, quadratus lumborum, gluteus medius and iliopsoas muscles as well as
the vertebral body at the level of the caudal endplates of L5, L6 and L7 (Figure 2).
Measurements were performed at the caudal endplates of L5, L6, and L7 to allow inclusion of
more muscles in the analyses and to account for variable conspicuity and anatomic variation in
size of muscles at various levels. For example, the quadratus lumborum muscle was best
visualized in the cranial images, and the gluteus medius muscle predominated at the L7 caudal
endplate. Muscle measurements were performed using a soft tissue window (window center (C)
= 40 Hounsfield units (HU); window width (W) = 400 HU) and bone measurements were
performed using a bone window (C = 400 HU; W = 1700 HU). Each measurement was made
twice by both observers, and the mean value for each was used for further calculations. Mean
CSA was determined between the left and right side for each muscle group, and was calculated
as a ratio to the CSA of the vertebral endplate within the same image slice. Asymmetry indices
(ASI) were calculated for each muscle group according to the following equations, employing
the method of Reeves et al. (2006):
𝑅𝑎𝑡𝑖𝑜 =𝑟𝑖𝑔ℎ𝑡 𝑚𝑢𝑙𝑡𝑖𝑓𝑖𝑑𝑢𝑠 𝐶𝑆𝐴
𝑙𝑒𝑓𝑡 𝑚𝑢𝑙𝑡𝑖𝑓𝑖𝑑𝑢𝑠 𝐶𝑆𝐴
𝐼𝑓 𝑟𝑎𝑡𝑖𝑜 ≥ 1, 𝑆𝑦𝑚𝑚𝑒𝑡𝑟𝑦 % = (𝑟𝑎𝑡𝑖𝑜 − 1) × 100
𝐼𝑓 𝑟𝑎𝑡𝑖𝑜 < 1, 𝑆𝑦𝑚𝑚𝑒𝑡𝑟𝑦 % = − ((1
𝑟𝑎𝑡𝑖𝑜) − 1) × 100
14
With this calculation, an asymmetry index value of zero would indicate perfect symmetry and a
value of 100 would indicate a two-fold difference in CSA between paraspinal muscles on each
side. Density in HU was also measured for a central region of interest within each muscle.
Exercise Protocol
Following VAS scoring, functional evaluation and CT scans, four dogs within the lumbosacral
pain group were enrolled in an 8-week core conditioning exercise program. The dogs were
selected as a convenience sampling, based on the projected training schedules. Dogs were
exercised for approximately 45 minutes per session, three times per week, consisting of four
progressive stages outlined in the exercise protocol (Appendix C). These progressive stages
were based on recommendations from the human literature for core strengthening exercises in
people with chronic low back pain (Danneels et al. 2000, Kim et al. 2011, Macedo et al. 2009,
O’Sullivan et al. 1997, Stuge et al. 2004) and knowledge of canine rehabilitation:
Weeks 1 and 2: Focus on isometric and light conditioning
Weeks 3 and 4: Increase strength/endurance at the level used in weeks 1-2
Weeks 5 and 6: Focus on controlled concentric and eccentric exercises, dynamic
mobilization and moderate conditioning
Weeks 7 and 8: Increase strength/endurance at the level used in weeks 5-6
The remaining four dogs classified as having lumbosacral pain did not participate in the exercise
program and served as controls for pain and function assessments. Activities for all dogs outside
15
of the exercise program were limited to brief slow, controlled leash walks, as regulated for all
locally housed military working dogs.
Within one week of completing the 8-week period of exercise, all dogs with lumbosacral pain
received repeated assessment including VAS score for lumbosacral pain, military working dog
functional assessment questionnaire, and computed tomography using the same technique as
previously described. The same individual that performed pain and functional assessment (S.M.)
or CT scans (P.G.) prior to the exercise intervention also performed the 8-week assessments to
eliminate inter-rater variability. The observers were unaware of treatment groups at the time of
the second observation. However, evaluators of the computed tomography images were not
blinded for the second CT scan assessment because only the exercised dogs received repeated
CT scans at the end of the 8-week program due to time and personnel constraints at the facility.
Data Analysis
An independent-samples t-test was used to compare age between the study populations, and
Pearson’s chi-squared analysis was used to compare breeds. Gender was not analyzed
statistically because an equal number of males and females were present in each group. Mean
goniometric tail extension angle (degrees) and lumbosacral dorso-ventral pressure (PSI)
collected at initial screening exams were compared between dogs with and without subjective
findings suggestive of lumbosacral pain. Independent-samples t-tests were performed to
compare objective measurements between pain and control populations in each case.
16
Mean values among the three slices for muscle asymmetry index, density and cross-sectional
area relative to the vertebra were compared between treatment groups for the five evaluated
muscle groups using independent-samples t tests.
Independent samples t-tests were used to compare mean muscle:vertebral CSA, asymmetry
indices and densities between the control population of dogs and those with lumbosacral pain.
For dogs that underwent the 8-week exercise protocol, paired samples t-tests were used to
compare mean pre- and post-exercise paraspinal muscle CSA, asymmetry and density values.
Equality of variance was evaluated using Levene’s Test for Equality of Variance. Statistical
analyses were performed using IBM SPSS® Statistics (IBM
®, Version 22, 2013, NY, USA).
Agreement between the two repeated measurements within observer (intra-observer reliability)
and between observers (inter-observer reliability) was assessed by evaluating the coefficient of
accuracy (Lin 1989) for mean muscle-to-L7 ratio, asymmetry index and density measurements.
Upon completion of all measurements, a board-certified radiologist (S.H.) evaluated the entire
image series of each patient to determine presence or absence of imaging evidence of
degenerative lumbosacral stenosis. Specific criteria evaluated were foraminal stenosis, dorso-
ventral narrowing of the spinal canal at the lumbosacral junction, loss of epidural fat, and
spondylosis deformans.
17
Chapter 3: Results
Case Selection
Forty-one military working dogs that met the inclusion criteria underwent initial orthopedic and
neurologic evaluations by the principal investigator. Twenty-four dogs with no orthopedic or
neurological findings (excluding obvious lumbosacral pain) were selected to proceed with 1)
pain and functional assessment by the assigned evaluator (S.M.) and 2) radiographic and CT
imaging. Two dogs were removed from the study between the initial screening and the imaging
procedures due to development of a clinical condition unrelated to the study. Additionally, three
dogs were eliminated based on pelvic radiographs; two due to evidence of hip osteoarthritis and
one due to presence of a transitional lumbosacral vertebra. Finally, three dogs were not included
in the data analysis because their VAS and/or functional questionnaire scores were equivocal,
preventing clear placement in either the control or lumbosacral pain study groups. A total of 16
military working dogs were therefore included in the study population; 8 dogs with and 8 dogs
without evidence of lumbosacral pain.
Study Populations and Initial LS Pain Assessment
The mean age of the control group was 6.6 years; that of the group with lumbosacral pain was
6.0 years. This difference was not significant (p = 0.437). There was no association between
breed and study group (p = 0.320). There were four males and four females in each group,
indicating no gender difference between groups overall. However, an unbalanced distribution of
18
0.00
5.00
10.00
15.00
20.00
25.00
Pre
ssu
re (
PSI
)
Control
LS Pain
neutered vs. intact males and females was present in that the control group contained the only
intact female and all control males were intact. All males in the LS pain group were castrated.
Goniometric tail angle and LS pressure data at initial screening were significantly different
between dogs found to have subjective evidence of lumbosacral pain (n = 11) and those that did
not (n = 13) at the initial screening (A.H.. Figure 4). Angle of dorsal tail extension was
significantly different between groups (p = 0.0002) at the initial screening, with the lumbosacral
pain group having reduced dorsal deviation. Mean goniometric angle of dogs with LS pain was
109.0 +/- 14.9; that of dogs without pain was 85.1 +/- 6.3. Pressure algometer readings were
also lower (p = 0.0035) for the group with LS pain. Mean tolerated LS pressure was 13.01 PSI
+/- 4.73 PSI for dogs with LS pain, and 18.65 PSI +/- 3.09 PSI for dogs without pain.
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
An
gle
of
Tail
Exte
nsi
on
(d
egr
ee
s)
Control
LS Pain
Figure 4: Differences between dogs with LS pain and control dogs regarding A,
goniometric angle of tail extension and B, dorso-ventral pressure over the L7-S1 disc
space. *p<0.001, ◊p <0.01
◊
*
A
B
19
Figure 5: Correlation between VAS (out of 100%) and functional questionnaire score
(out of 32 possible points) at the baseline pain and function evaluation
R² = 0.6704
0%
10%
20%
30%
40%
50%
60%
0 5 10 15 20 25
VA
S Sc
ore
(%
)
Functional Questionnaire Score
All data for comparison of outcome measures were found to have equal variance by Levene’s
Test for Inequality of Variance. Mean VAS scores (maximum possible pain = 100%) were 34%
for dogs with lumbosacral pain and 1% for control dogs. Mean functional questionnaire scores
(maximum dysfunction = 32) were 15.88 for dogs with lumbosacral pain and 0.63 for dogs in the
control group. The differences between groups were statistically significant (p < 0.001) for both
VAS and functional questionnaire scores. Furthermore, there was a moderate to strong positive
linear association between VAS pain level and functional disability as determined by the
functional questionnaire (r2 = 0.670, Figure 5). There was no significant difference between
exercised and non-exercised dogs within the LS pain group for VAS or functional questionnaire
score at the baseline evaluation (p = 0.093 and p=0.412, respectively).
20
Lumbar Paraspinal Muscle CSA, Asymmetry Index and Density Pre-Exercise
All data for comparison of muscle parameters were found to have equal variance by Levene’s
Test for Inequality of Variance. Mean muscle-to-vertebral CSA ratios (Figure 6) for the
multifidus lumborum and longissimus lumborum were significantly smaller in dogs with LS pain
than in control dogs (p = 0.020 and p = 0.021, respectively). The mean muscle-to-vertebral CSA
ratio for the epaxial muscles combined (multifidus lumborum and longissimus lumborum) was
also significantly smaller in dogs with LS pain than in control dogs (p = 0.0095). The mean
muscle-to-vertebral CSA ratio for gluteus medius was significantly larger in dogs with LS pain
than in those without (p = 0.043). There were no significant differences between the study
populations for muscle:vertebral CSA ratio for any other muscle group. Left and right multifidus
lumborum ASI (Figure 7) was significantly higher in dogs with LS pain than in those without at
the baseline evaluation (p = 0.0005). Control dogs demonstrated a higher ASI for the gluteus
medius muscle than dogs with LS pain (p = 0.011). There were no other significant differences
in muscle ASI between LS pain and control groups. Mean muscle density values were higher in
control dogs than in dogs with LS pain only for the multifidus lumborum (p = 0.03) and
quadratus lumborum (p = 0.011) muscle groups (Figure 8).
21
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
MF LL MF + LL QL IPS GM
Mu
scle
:Ve
rte
bra
l CSA
Rat
io
LS Pain
Control
◊
0.00
5.00
10.00
15.00
20.00
25.00
30.00
MF LL QL IPS GM
ASI
(0
ind
icat
es
pe
rfe
ct s
ymm
etr
y)
LS Pain
Control
*
◊
*
¥
Figure 7: Baseline muscle ASI for dogs with LS pain and control dogs. An ASI value of zero
indicates perfect symmetry; 100 would indicate that muscle on one side would have twice the
CSA of the muscle on the opposite side. *,◊ p<0.05. Error bars represent one standard
deviation above and below the mean.
Figure 6: Baseline muscle:vertebral CSA ratios for dogs with LS pain and control dogs
*,◊ p<0.05, ¥ p <0.01. MF = Multifidus lumborum, LL = longissimus lumborum, QL =
Quadratus lumborum, IPS = Iliopsoas, GM = Gluteus medius. Error bars represent one
standard deviation above and below the mean.
22
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
MF LL QL IPS GM
Mu
scle
De
nsi
ty (
HU
)
LS Pain
Control
* ◊
Exercise Protocol
The four dogs with lumbosacral pain that received core conditioning were able to perform all
exercises and complete the 8-week program. Based on detailed subjective observations made
by the individual administering the exercises, dogs demonstrated a tendency to have mild to
moderate difficulty with some exercises at the beginning of each new stage in the bimonthly
progression, with improvement by the end of the two weeks. No dog in either the exercised or
rested group required rescue analgesia nor received a change in activity throughout the study.
Figure 8: Baseline muscle density (HU) for dogs with LS pain and control dogs
*,◊ p<0.05. Error bars represent one standard deviation above and below the mean.
23
Figure 9: Functional questionnaire (A) and VAS (B) scores for military working dogs with
mild lumbosacral pain before and after exercise. * p<0.05
Pain and Functional Assessments: Post-Exercise
The mean percent change in VAS score after the 8-week exercise protocol was -9% for dogs that
exercised and was -31% for dogs that did not undergo exercise. However, the change in VAS
scores after the 8-week study protocol was not significant in either the exercised or rested dogs
(p=0.379 and p=0.066, respectively, Figure 9). The mean percent change in functional
questionnaire score was -57% in dogs that performed the exercises and was -6% in dogs that did
not exercise. Functional questionnaire scores improved significantly in dogs that were exercised
(p=0.031), but not in those that were rested (p=0.828), between the evaluation time points.
There was no association between VAS pain level and functional disability score (r2
= 0.008) at
the post-exercise evaluation of dogs in either group.
0
5
10
15
20
25
30
Pre-Exercise Post-Exercise
Qu
est
ion
nai
re S
core
Exercised Group
Rested Group
*
*
0%
10%
20%
30%
40%
50%
60%
Pre-Exercise Post-Exercise
VA
S Sc
ore
(%
)
Exercised Group
Rested Group
A B
24
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
MF LL
Mu
scle
:Ve
rte
bra
l CSA
Rat
io
0
5
10
15
20
25
MF LL
Mu
scle
ASI
Pre-Exercise
Post-Exercise
B
40
42
44
46
48
50
52
54
56
58
60
MF LL
Mu
scle
De
nsi
ty (
HU
)
Lumbar Paraspinal Muscle CSA, Asymmetry Index and Density Post-Exercise
Dogs with lumbosacral pain that received exercise had a significantly increased muscle:vertebral
CSA ratio for the multifidus lumborum muscle (p = 0.019), but ratios were not significantly
different after exercise for any other muscle group assessed (Figure 10). Additionally, exercised
dogs demonstrated a significant decrease in asymmetry index (p = 0.002) and a significant
increase in density (p = 0.024) for the multifidus lumborum muscles upon evaluation of CT
scans after the 8-week exercise protocol. ASI and density did not significantly change after
exercise for any other muscle group assessed.
Figure 10: Pre- and post-exercise multifidus lumborum and longissimus lumborum
muscle:vertebral CSA ratio (A), ASI (B) and density (C) for dogs with LS pain that underwent
the 8-week exercise program. * p < 0.05. Error bars represent one standard deviation above
and below the mean.
A
C *
*
*
25
Intra- and Inter-Observer Agreement
Random sampling of ten measurements to assess agreement between the two observers for
muscle CSA measurements demonstrated a concordance correlation coefficient of 0.9868,
demonstrating excellent agreement. Moderate agreement (CCC = 0.7988) was seen for muscle
density measurements. No agreement (CCC = 0.1491) was noted for ASI measurements.
Concordance correlation coefficient for intra-rater repeatability between two measurements
demonstrated substantial agreement for muscle CSA (CCC = 0.9965) and density (CCC = 0.988)
for the observer A.H. For the observer S.H., substantial agreement was present with respect to
cross sectional area (CCC = 0.9981), with moderate agreement regarding measurements of
muscle density (CCC = 0.8814).
Imaging and Clinical Agreement
There was no association between clinical findings of lumbosacral pain and CT-based structural
findings indicative of degenerative lumbosacral stenosis. Only two of the sixteen dogs had no
evidence of DLSS on CT imaging evaluation. One dog had been categorized as a control and
one had been categorized as having lumbosacral pain. Five dogs had equivocal evidence of
DLSS on CT imaging, characterized as foraminal stenosis or spondylosis deformans at L7-S1
without other accompanying structural changes or obvious nerve root compression. Three of
these dogs had been categorized as having lumbosacral pain; two had been categorized as
controls based on clinical and functional assessment. The remaining nine dogs were diagnosed
26
as having DLSS based on imaging. Four of these were dogs with lumbosacral pain and five had
been clinically and functionally determined to be controls.
27
Chapter 4: Discussion
Both Belgian Malinois and German Shepherd Dogs are the two most common breeds in the U.S.
Military Working Dog program. German Shepherd Dogs have demonstrated a substantial
predisposition for lumbosacral stenosis (Daniellson & Sjostrum 1999, De Risio et al. 2001,
Suwankong et al. 2008). Although this study investigated a small number of dogs, distribution
of breeds between study groups was suggestive that German Shepherd Dogs and Belgian
Malinois may be similarly predisposed to lumbosacral stenosis. This is supported by
identification of radiographic/computed tomographic indicators of lumbosacral disease even in
several clinically normal military working dogs of both breeds. These findings warrant further
study and emphasize a need to investigate improved diagnostic and management strategies for
DLSS in military working dogs.
Lack of agreement between CT characteristics of DLSS and lumbosacral pain supported findings
by Jones et al. (2000) for working dogs with DLSS treated surgically, in which clinical findings
were more important predictors of post-operative outcome than imaging evidence. Additionally,
symptoms in human patients with CLBP have a poor association with diagnostic imaging
findings (Beattie et al. 2000, Takatalo et al. 2011). These findings indicate a need for more
objective pain or functional assessment tools for dogs with lumbosacral pain. The screening
evaluation in this study provided preliminary evidence (Appendix) in support of two objective
assessment tools: goniometric tail angle and dorso-ventral pressure with an algometer over L7-
S1. Outcomes from these objective measures were significantly different between the group of
dogs that had pain and those that did not by subjective evaluation at the same time. However,
28
subjective evaluation and objective data collection for goniometry and tail elevation angle were
performed by the same individual, prohibiting blinding.
Three dogs that did not demonstrate subjective signs of LS pain at initial screening were
determined to have lumbosacral pain at the follow-up evaluation based on VAS scores and
functional questionnaire assessment. One dog identified as having LS pain and one dog with
questionable pain at initial screening exam were re-classified as controls at the follow-up
evaluation based on VAS and functional assessment tests. This difference in findings may be
due to the variable nature of pain, or to differences in criteria of assessment used by two separate
individuals. Having two distinct observers at two different time-points was a logistical necessity
that posed a limitation to the study design. Comparison of goniometric angle of tail elevation
and objective dorso-ventral pressure at L7-S1 to other means of assessment by a single
individual may provide more definitive information about their usefulness as outcome measures
for DLSS.
Human literature has demonstrated statistically significant findings for reduction of low back
pain with exercise with as few as seven subjects. Hides et al. (2008) found that specific core
stabilization exercises for seven cricket athletes with low back pain reduced their pain by 50%
when compared to pain before exercise. Based on this difference, a power analysis was
performed on the current study population and demonstrated that a difference of 50% could be
detected if 10 dogs were exercised. However, only four dogs could logistically be included in
the exercise program in this study due to availability of personnel and availability of dogs to
complete the 8-week exercise protocol. Baseline VAS scores in the LS pain dogs did not differ
29
significantly between the group that underwent exercise and that which did not. VAS scores
improved for all subjects by the end of the study period. Although the change was not
significant for either group, there was a trend toward greater change favoring reduction in pain in
the group that did not undergo the exercise program. Functional scores significantly improved in
the exercise group, while control group functional scores did not improve. Furthermore, VAS
scores and function scores correlated poorly at the second evaluation. This lack of objective
evidence of improved pain in the face of increased function may be due to the low number of
subjects used, or may be influenced by limitations inherent in the VAS scoring system.
Additionally, these results may suggest that while rest may improve pain, function does not
improve. Exercise, however, may improve function and pain (or prevent worsening of pain) in
dogs with DLSS. Because of the activities military working dogs must perform, our results
suggest that core strengthening exercises may significantly improve function without worsening
pain.
Recommendations from the Initiative on Methods, Measurement, and Pain Assessment in
Clinical Trials (IMMPACT) consensus meeting in 2008 (Dworkin et al.) regarding application of
outcome measures to human patients with CLBP included using at least two outcome measures
and understanding that baseline pain would affect interpretation of a change in VAS score, as all
changes of the same magnitude on the scale may not be clinically equivalent. The dogs that did
not undergo exercise in this study had higher baseline VAS scores than those that exercised, an
unexpected finding as groups were selected based on class training schedules. Although this
difference was not statistically significant it may have an impact on interpretation of the change
in scores after the 8-week study period. Hielm-Björkman et al. (2011) demonstrated poor face
30
validity with owners and VAS scores for osteoarthritis (OA) pain until the pain was alleviated
and subsequently reoccurred. It was postulated that a lack of owner experience in recognizing
specific signs of pain was the source. VAS for lameness was found to correlate poorly with
force plate gait analysis for dogs that were sound, mildly or moderately lame, and only showed
good correlation with vertical impulse for each observer if very lame dogs were included (Quinn
et al. 2007). The present study attempted to improve the accuracy of the VAS scoring system by
providing four criteria to consider in the assessment; however individual VAS scores for each
criterion may have demonstrated better reliability. Hudson et al. (2004) evaluated the
repeatability of owner-assessed VAS questionnaires in dogs with pain and lameness, using force
plate gait analysis as the gold standard for comparison. The 2004 study generated a 39-point
questionnaire addressing various signs of pain and dysfunction with a VAS score assigned to
each question. Forty-nine percent of the questions were found to have good repeatability with
Spearman rank correlation coefficient of 0.68-0.90. In the present study, the evaluator may have
ranked the criteria equally for the VAS score (i.e. dorsal tail elevation and hyperextension of the
spine) when there may be unequal distribution among such clinical signs within a population of
dogs with lumbosacral pain. Suwankong et al. (2008) found via retrospective evaluation that
only 5/156 pet dogs (3.2%) with DLSS demonstrated pain on tail extension, whereas pain on
hyperextension of the lumbar spine was observed in 107/156 (68%). Another retrospective study
found that 97.7% of 131 client-owned dogs diagnosed with DLSS had pain on hyperextension of
the lumbar spine and/or tail combined (Danielsson et al. 1999). The use of a single VAS scoring
system may be an insufficient tool to characterize lumbosacral pain, unless a questionnaire could
be developed similar to that used by Hudson et al. involving multiple specific criteria with a
separate VAS scoring system for each.
31
The functional assessment questionnaire used in this study was developed specifically to
evaluate the performance of military working dogs during familiar tasks used in search and
detection that are often sources of handler complaints when dogs begin to show signs that lead to
a subsequent diagnosis of DLSS. Therefore, there is no validation history for this questionnaire
as an objective assessment tool for evaluating dysfunction associated with lumbosacral pain. At
the baseline evaluation, function had a moderate to strong association with VAS score in dogs
with lumbosacral pain. This association was not apparent at the follow-up evaluation after the
exercise period, suggesting that mild pain may not significantly hinder function if appropriate
exercise protocols are instituted. While further studies need to be performed to assess internal
and external validity of the functional questionnaire, it appears to have promise as a tool for
evaluating military working dogs with lumbosacral pain trained in specific activities.
Multifidus lumborum and longissimus lumborum CSA ratios were significantly smaller in
military working dogs with lumbosacral pain than in control dogs. Additionally, multifidus
lumborum muscles had decreased density and increased asymmetry between the left and right
sides in dogs with lumbosacral pain. The other muscles evaluated had inconsistent or
insignificant differences in response to lumbosacral pain and exercise, and their role in
stabilization of the lumbosacral region in dogs remains unclear. This study corroborates the
human literature in which paraspinal atrophy occurs in people with CLBP and is often most
pronounced in the multifidus lumborum muscles (Danneels et al. 2000, Hides et al. 2008, Kamaz
et al. 2007). This study also supports the findings of an earlier pilot study performed
retrospectively by the principle investigator (Henderson et al. 2014). In this study, transverse
32
magnetic resonance (MR) images through the L7 caudal endplate were evaluated in nine dogs
with a diagnosis of LS stenosis or cauda equina syndrome and nine control dogs. Mean cross-
sectional area and symmetry of lumbar multifidus and longissimus lumborum muscles were
compared between the two groups. There were significant differences between dogs with DLSS
and control dogs in muscle CSA to L7 vertebral endplate CSA ratio means for both muscles (p =
0.027 for multifidus lumborum, p = 0.011 for longissimus lumborum). Mean muscle asymmetry
indices were higher in dogs with DLSS, but the differences were not statistically significant. In
accordance with these pilot study findings, power analysis (α set at 0.05, β set at 0.80) suggested
that ten dogs identified as having lumbosacral pain according to the parameters above were
needed to detect a significant difference between groups for paraspinal muscle cross-sectional
area. Only eight dogs per group met all inclusion criteria for the present study, yet statistical
significance was achieved. One difference between the pilot MRI study and the present study
was that breed was not controlled in the former, resulting in an over-representation of German
Shepherd Dogs in the group with lumbosacral stenosis. Although vertebral cross-sectional area
of German Shepherd Dogs was similar to that of other breeds, the role of breed-related
conformation in the muscle:vertebral area ratio could not be eliminated. The present study
demonstrated, however, that multifidus lumborum and longissimus lumborum atrophy were
present in dogs with lumbosacral pain in comparison to structurally similar dogs without pain.
This suggests that differences in muscle atrophy and asymmetry are associated with pain and not
with conformational characteristics of the dogs.
Another characteristic unique to the present study was the inclusion of paraspinal muscle density
measurements. In humans, sarcopenia (age-related muscle loss in the absence of diagnosed
33
disease) results in approximately 30% loss of muscle mass from 20 to 80 years of age (Freeman
2012). Similar investigations in veterinary patients are limited, but initial studies have
demonstrated significant loss of muscle mass with age in healthy dogs (Freeman 2012, Lawler et
al. 2009). Age was not different between control and LS pain groups in the present study.
Increased paraspinal muscle fat content has also been observed in people with CLBP when
compared to asymptomatic individuals (Bouche et al. 2011, Parkkola et al. 1993). Fat has a
lower density (HU) than muscle when evaluated by computed tomography. Increased
intramuscular fat may have accounted for the decreased density of the multifidus lumborum in
dogs with lumbosacral pain in this study, and muscle hypertrophy may have explained the
increased density seen in the same muscle group in response to exercise. Additional studies
involving precise algorithms for normalizing fat content are warranted to further investigate the
potential effect of lumbosacral pain and exercise on intramuscular fat content in dogs.
For people with CLBP, paraspinal musculature, pain and functional response to exercise
programs are variable in the literature. Several mechanisms for CLBP have been proposed for
people, calling for a classification system that may further guide diagnosis and therapeutic
intervention (Gudavalli et al. 2006, Maluf et al. 2000, O’Sullivan 2005). O’Sullivan has
postulated that there is a large subcategory of individuals with chronic low back pain that have
ongoing symptoms secondary to compensation for a mechanical deficit or excess of stability. He
argues that these patients may be more responsive to management with therapeutic exercise
regimens designed to alter paraspinal muscle activation patterns. This could be a common
mechanism for the pain associated with DLSS in dogs as well. Therefore, dogs with structural
abnormalities at the lumbosacral junction may be able to respond favorably to a similar exercise
34
regimen. Various exercise protocols recommended to improve pain, function and increase
paraspinal muscle area in humans include beginning with sustained low-load contractions, then
gradually adding in limb movements mimicking functional tasks, progressively increasing in
difficulty (O’Sullivan et al. 1997). Additionally, studies with humans and CLBP have found that
a combination of stabilization and resistance exercises is necessary to elicit significant increases
in paraspinal cross-sectional area (Kim et al. 2011, Danneels et al. 2000). We found no
published canine study that identified significant changes in paraspinal muscles in response to
exercise. However, a study by Stubbs et al. (2011) compared pre- and post- dynamic
mobilization exercise effects on thoracolumbar multifidus muscle cross-sectional area in eight
healthy horses. All subjects demonstrated a statistically significant increase in muscle mass after
exercise (p < 0.05). However, the study did not examine the effect of exercises on
lumbar/lumbosacral pain level in an affected population. The present study demonstrated a
significant (p < 0.05) increase in multifidus lumborum cross-sectional area, symmetry and
density in military working dogs with lumbosacral pain in response to core-strengthening. These
findings would ideally have been compared to repeated CT scans of dogs with pain that were
rested. However, there were insufficient resources to repeat CT scans, which imposed a
limitation on the experimental design.
Results of this study indicate that functional improvement for tasks required during detection
duties and an increase in size of the lumbar multifidus muscles can be achieved with a core
conditioning program in dogs with mild to moderate lumbosacral pain without neurological
deficits. Additional research to assess for a consistent and sustained response may involve
modeling the exercise program more closely after those found most successful in the human
35
CLBP patient population. Future investigation of effects of lumbosacral pain and core
strengthening on paraspinal muscles would ideally include electromyography to compare muscle
activation patterns between control groups and treated groups undergoing exercise and training.
To determine whether increase in muscle size in response to exercise may be more profound and
include additional muscle groups, application of a 10-12 week exercise program with additional
resistance exercises should be evaluated. Further work in dogs with more severe pain could
result in greater differences in pain, function, paraspinal muscle area and muscle density in
response to exercise. However, the case population used in this study was intended to represent
that which would be most likely to return to duty with a conservative physical rehabilitation
program as the initial treatment intervention.
In this study, Military Working Dogs with clinically significant lumbosacral pain had increased
cross-sectional area, density and symmetry in multifidus lumborum as well as improved function
in response to an 8-week core strengthening program. This is the first study to evaluate
muscular, functional and pain response of dogs with lumbosacral pain to any exercise regimen.
A conservative, evidence-based physical rehabilitation program to address lumbosacral pain
would be highly worthwhile to the DoD Military Working Dog Program since the cost of
training new military working dogs to replace those no longer able to perform exceeds $30,000
per animal. Long-term possible benefits may include the development of a paraspinal muscle
conditioning program that could be provided to military working dogs as an aid to prevent
lumbosacral pain in otherwise predisposed dogs.
36
References
37
Barr, K. P., Griggs, M., Cadby, T. (2005) Lumbar stabilization: Core concepts and current
literature, part 1. American Journal of Physical Medicine and Rehabilitation 84, 473-480
Beattie, P. F., Meyers, S. P., Stratford, P. (2000) Associations between patient report of
symptoms and anatomic impairment visible on lumbar magnetic resonance imaging. Spine
25, 819-828
Benninger, M. I., Seiler, G. S., Robinson, L. E., et al.. (2006) Effects of anatomic
conformation on three-dimensional motion of the caudal lumbar and lumbosacral portions of
the vertebral column of dogs. American Journal of Veterinary Research 67, 43-50
Bouche, K. G. W., Vanovermeire, O., Stevens, V. K. (2011) Computed tomographic analysis
of the quality of trunk muscles in asymptomatic and symptomatic lumbar discectomy
patients. BioMed Central Musculoskeletal Disorders 12, 1-9
Carneiro, K.A., Rittenberg, J.D. (2010) The role of exercise and alternative treatments for
low back pain. Physical Medicine and Rehabilitation Clinics of North America 21, 777-792
Danielsson, F. & Sjostrom, L. (1999) Surgical treatment of degenerative lumbosacral
stenosis in dogs. Veterinary Surgery 28, 91-98
Danneels, L. A., Vanderstraeten, G. G., Cambier, D. C., et al.. (2000) CT imaging of trunk
muscles in chronic low back pain patients and healthy control subjects. European Spine
Journal 9, 266-72
38
De Risio, L., Sharp, N. J., Olby, N. (2001) Predictors of outcome after dorsal decompressive
laminectomy for degenerative lumbosacral stenosis in dogs: 69 cases (1987–1997). Journal
of the American Veterinary Medical Association 219, 624-628
Dworkin, R. H., Turk, D. C., Wyrwich, K. W., et al. (2008) Interpreting the Clinical
Importance of Treatment Outcomes in Chronic Pain Clinical Trials: IMMPACT
Recommendations. The Journal of Pain 9, 105-121
Ebenbichler, G. R., Oddsson, L. I., Kollmitzer, J. et al. (2001) Sensorymotor control of the
lower back: Implications for rehabilitation. Medicine and Science in Sports and Exercise 33,
1889–1898
Freeman, L. M. (2012) Cachexia and sarcopenia: Emerging syndromes of importance in dogs
and cats. Journal of Veterinary Internal Medicine 26, 3-17
Gibbons, L., Videman, T., Battie, M. (1997) Isokinetic and psychophysical lifting strength,
static back muscle endurance, and magnetic resonance imaging of the paraspinal muscles as
predictors of low back pain in men. Scandinavian Journal of Rehabilitation Medicine 29,
187-191
Gudavalli, M. R., Cambron, J. A., McGregor, M., et al. (2006) A randomized clinical trial
and subgroup analysis to compare flexion–distraction with active exercise for chronic low
back pain. European Spine Journal 15, 1070–1082
Henderson, A. L., Millis, D. L., Hecht, S., et al. (2014) Cross-sectional area and asymmetry
of lumbar paraspinal muscles for dogs with and without degenerative lumbosacral stenosis.
39
Proceedings of the 4th World Veterinary Orthopedic Congress and 41st Veterinary Orthopedic
Society Conference, March 1 to 8, Breckenridge, CO. Veterinary Orthopaedics and
Traumatology p. A1
Hermanson, J. W. & Evans, H. E. (1993) Muscles of the pelvic limb, in:Evans, H. E. (ed):
Miller’s Anatomy of the Dog. 3rd ed. Ed H. E. Evans. W.B. Saunders, Philadelphia. pp 258-
384
Hides, J. A., Stanton, W. R., McMahon, S., et al. (2008) Effect of stabilization training on
multifidus muscle cross-sectional area among young elite cricketers with low back pain.
Journal of Orthopaedic and Sports Physical Therapy 38, 101-108
Hielm-Bjorkman, A. K., Kapatkin, A. S., Rita, H. J. (2011) Reliability and validity of a
visual analogue scale used by owners to measure chronic pain attributable to osteoarthritis in
their dogs. American Journal of Veterinary Research 72, 601-607
Hodges, P. (2004) Lumbopelvic stability: a functional model of the biomechanics and motor
control, in: Richardson C (ed): Therapeutic Exercise for Lumbopelvic Stabilization, 2nd
ed.
Edinburgh, Churchill Livingstone.pp 13–28
Hudson, J. T., Slater, M. R., Taylor, L. (2004) Assessing repeatability and validity of a
visual analogue scale questionnaire. Journal of Veterinary Research 65, 1634-1643
Jones, J. C., Banfield, C. M., Ward, D. L. (2000) Association between postoperative outcome
and results of magnetic resonance imaging and computed tomography in working dogs with
degenerative lumbosacral stenosis. Journal of the American Veterinary Medical Association
216, 1769-1774
40
Kader, D. F., Wardlaw, D., Smith, F. W. (2000) Correlation between the MRI changes in the
lumbar multifidus muscles and leg pain. Clinical Radiology 55, 145-149
Kamaz, M., Kireşi, D., Oğuz, H., et al.. (2007) CT measurement of trunk muscle areas in
patients with chronic low back pain. Diagnostic and Interventional Radiology 13, 144-148
Kim, J., Gong, W., Hwang, B. (2011) The effects of resistivity and stability-combined
exercise for lumbar muscles on strength, cross-sectional area and balance ability: exercises
for prevention of lower back pain. Journal of Phyical Therapy Science 23, 247-50
Lawler, D. F., Larson, B. T., Ballam, J. M., et al. (2009) Diet restriction and ageing in the
dog: Major observations over two decades. British Journal of Nutrition 9, 793-805
Lin, L. I. (1989) A concordance correlation coefficient to evaluate reproducibility.
Biometrics 45, 255-268
Linn, L. L., Bartels, K. E., Rochat, M. C., et al. (2003) Lumbosacral stenosis in 29 military
working dogs: epidemiologic findings and outcome after surgical intervention (1990-1999).
Veterinary Surgery 32, 21-29
Maluf, K. S., Sahrmann, S. A., Van Dillen, L. R. (2000) Use of a classification system to
guide nonsurgical management of a patient with chronic low back pain. Physical Therapy 80,
1097–1111
Marshall, P.W., Desai, I., Robbins, D.W.. (2011) Core stability exercises in individuals with
and without chronic nonspecific low back pain. Journal of Strength and Conditioning
Research 25, 3404-3411
41
Marshall, P. W. & Murphy, B. A. (2006) Evaluation of functional and neuromuscular
changes after exercise rehabilitation for low back pain using a Swiss ball: a pilot study.
Journal of Manipulative and Physiological Therapeutics 29, 550-560
Meij, B. P., Bergknut, N. B. (2010) Degenerative lumbosacral stenosis in dogs. Veterinary
Clinics of North America: Small Animal Practice 40, 983-1009
Moore, G. E., Burkman, K. D., Carter, M. N., et al. (2001) Causes of death or reasons for
euthanasia in military working dogs: 927 cases (1993-1996). Journal of the American
Veterinary Medical Association 219, 209-214
Niemistö, L., Lahtinen-Suopanki, T., Rissanen, P., Lindgren, K.A., Sarna, S., Hurri, H.
(2003) A randomized trial of combined manipulation, stabilizing exercises, and physician
consultation compared to physician consultation alone for chronic low back pain. Spine 28,
2185-2191
O’Sullivan, P. B. (2005) Diagnosis and classification of chronic low back pain disorders:
maladaptive movement and motor control impairments as underlying mechanism. Manual
Therapy 10, 242-255
O'Sullivan, P. B., Phyty, G. D., Twomey, L. T., et al. (1997) Evaluation of specific
stabilizing exercise in the treatment of chronic low back pain with radiologic diagnosis of
spondylolysis or spondylolisthesis. Spine 22, 2959-2967
42
O'Sullivan, P. B., Twomey, L., Allison, G. T. (1998) Altered abdominal muscle recruitment
in patients with chronic back pain following a specific exercise intervention. Journal of
Orthopedics and Sports Physical Therapy 27, 114-124
Panjabi, M. M. (1992) The stabilizing system of the spine. Part I. Function, dysfunction,
adaptation, and enhancement. Journal of Spinal Disorders 5, 383-389
Panjabi, M. M. (1992) The stabilizing system of the spine. Part II. Neutral zone and
instability hypothesis. Journal of Spinal Disorders 5, 390-397
Parkkola, R., Rytökoski, U., Kormano, M. (1993) Magnetic resonance imaging of the discs
and trunk muscles in patients with chronic low back pain and healthy control subjects. Spine
1, 830-836
Quinn, M. M., Keuler, N. S., Lu, Y., et al. (2007) Evaluation of agreement between
numerical rating scales, visual analogue scoring scales, and force plate gait analysis in dogs.
Veterinary Surgery 36, 360-367
Reeves, N.P., Cholewicki, J., Silfies, S.P. (2006) Muscle activation imbalance and low-back
injury in varsity athletes. Journal of Electromyography and Kinesiology. 16, 264-272
Renkawitz, T., Boluki, D., Grifka, J. (2006) The association of low back pain, neuromuscular
imbalance, and trunk extension strength in athletes. The Spine Journal 6, 673-683
Smit, T. H. (2002) The use of a quadruped as an in vivo model for the study of the spine –
biomechanical considerations. European Spine Journal 11, 137-144
43
Stubbs, N. C., Kaiser, L. J., Hauptman, J., et al. (2011) Dynamic mobilisation exercises
increase cross sectional area of musculus multifidus. Equine Veterinary Journal 43, 522-52
Stuge B, Veierod MB, Laemm E, et al. (2004) The efficacy of a treatment program focusing
on specific stabilizing exercises for pelvic girdle pain after pregnancy: a two-year follow-up
of a randomized clinical trial. Spine 29, E197-E203
Suwankong, N., Meij, B. P., Voorhout, G., et al. (2008) Review and retrospective analysis of
degenerative lumbosacral stenosis in 156 dogs treated by dorsal laminectomy. Veterinary
and Comparative Orthopaedics and Traumatology 21, 285-293
Takatalo, J., Karppinen, J., Niinimäki J. et al. (2011) Does lumbar disc degeneration on
magnetic resonance imaging associate with low back symptom severity in young Finnish
adults? Spine 36, 2180-2189
Teeling, M. & Van den Berg, S. S. (2012) Exploring the effects of canine core ball exercises
on the multifidus muscles and hind quarter muscles in three dogs over two months.
Proceedings of the 7th International Symposium in Veterinary Rehabilitation and Physical
Therapy. August 15 to 18, Vienna, Austria. p. 100
Van Dieën, J.H., Kingma, I., Van der Bug, P. (2003) Evidence for a role of antagonistic
cocontraction in controlling trunk stiffness during lifting. Journal of Biomechanics 36, 1829-
1836
Zimmerman, M. C., Vuono-Hawkins, M., Parsons, J. R., et al. (1992) The mechanical
properties of the canine lumbar disc and motion segment. Spine 17, 213-220
44
Appendices
45
Appendix A: Orthopedic and Neurologic Examination
Neurological Examination:
OBSERVATIONS
Behavior Normal Aggressive Excited Anxious Apathetic
Gait Ataxia Circling Hypometria Hypermetria Lameness Affected
Limbs:
Voluntary
Movement Normal
Monoparesis
/plegia
Paraparesis
/plegia
Tetraparesis
/plegia Tremors
LF RF LH
RH
POSTURAL
REACTIONS
LIMB
AND DEGREE KEY
Hopping LF RF LH RH 0 = absent
Hemiwalking LF RF LH RH +1 = decreased
Proprioceptive
Positioning (Knuckling) LF RF LH RH +2 = normal
Extensor Postural
Thrust LF RF LH RH +3 = exaggerated
Placing: Tactle LF RF LH RH
Placing: Visual LF RF LH RH
REFLEXES LIMB AND DEGREE
KEY
MYOTATIC Pelvic Thoracic
0 = absent
Left Patellar Gastroc Cranial
Tibialis Triceps Biceps ECR +1 = decreased
Right Patellar Gastroc Cranial
Tibialis Triceps Biceps ECR +2 = normal
Flexion LF RF LH RH Perineal Cutaneous
Trunci +3 = exaggerated
Crossed Extensor LF RF LH RH +4 = clonus
ADDITIONAL
INFORMATION
Hyperesthesia Cervical? Y/N Thoracolumbar
Y/N Location/Comments:
Bladder Control Normal Abnormal Comments:
Bowel Control Normal Abnormal Comments:
46
Orthopedic Examination:
GAIT EVALUATION LEFT THORACIC RIGHT
THORACIC LEFT PELVIC RIGHT PELVIC
Lameness at a Stance 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
Lameness at a Walk 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Lameness at a Trot 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Zero is no lameness, 4 (at a stance) and 5 (at a walk/trot) represent continuous non-weight-bearing lameness
ADDITIONAL
INFORMATION PALPATION
JOINT
RANGE OF
MOTION
ATROPHY
Right Thoracic Pain? Y / N
Asymmetry: Y / N
Restricted? Y / N
Painful? Y / N
None Mild
Mod Severe Comments/Location(s):
Left Thoracic Pain? Y / N
Asymmetry: Y / N
Restricted? Y / N
Painful? Y / N None Mild
Mod Severe Comments/Location(s):
Right Pelvic Pain? Y / N
Asymmetry: Y / N
Restricted? Y / N
Painful? Y / N None Mild
Mod Severe Comments/Location(s):
Left Pelvic Pain? Y / N
Asymmetry: Y / N
Restricted? Y / N
Painful? Y / N None Mild
Mod Severe Comments/Location(s):
ADDITIONAL FINDINGS/COMMENTS:-
_____________________________________________________________________________________
_____________________________________________________________________________________
_____________________________________________________________________________________
_____________________________________________________________________________________
________________________________________________________________________________
TEST PAIN TEST RESULTS
Lumbar
Hyperextension with
Hips in Flexion
Present Absent Algometer at L7-S1
(always after digital test) Pressure (PSI):
Dorsal Tail-Base
Elevation Present Absent
Goniometric Angle of
Max Tail Elevation (0
degrees is parallel to the
ground surface)
Angle (Degrees):
Digital Pressure over
L7-S1 Present Absent
Dog Meets Inclusion Criteria | Dog Does Not Meet Inclusion Criteria
LS Pain Present or LS Pain Absent
________________________
Signature of Examiner
47
Appendix B: Functional Assessment Questionnaire
Handler: First Name______________ Last Name_____________________
Branch of Service (circle one): Army Air Force Navy Marines
Dog: Name_________________ Sex (Circle One): Male Female
Spayed/Neutered? (Circle One): Yes No
Age (round to the nearest year)_______
Breed: German Shepherd Belgian Malinois
Functional Assessment
The following questions are about your evaluation of your military working dog’s performance during a
training session. Please read the following questions and answer each one to the best of your ability.
Provide only one answer for each question.
1. How long does the dog stay in the “hup” position during the evaluation routine?
a. 0 seconds, the dog refuses to “hup” when I ask
b. 1-2 seconds
c. 3-4 seconds
d. 5-6 seconds
e. 7 or more seconds
48
2. Does the dog show any difficulty “hupping” each time you ask, to a distance of at
least three feet off the ground at the dog’s shoulder level? Difficulty may be defined
as hesitation in going into the hup, repeatedly unsuccessful attempts at rising into the
“hup,” refusal to “hup” or crying out when performing the activity.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
3. Does the dog show any difficulty in jumping INTO a vehicle? Difficulty may be
defined as hesitation, inability to complete the jump the first time, crying out when
performing the activity, or slipping or falling during or after the jump.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
4. Does the dog show any difficulty in jumping OUT OF a vehicle? Difficulty may be
defined as hesitation, crying out when performing the activity, or slipping or falling
during or after the jump.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
49
5. Does the dog show any difficulty in climbing the staircase on an obstacle course?
Difficulty may be defined as hesitation, repeatedly attempting to get down, refusal to
climb, crying out when performing the activity, or slipping or falling during the climb.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
6. Does the dog show any difficulty in climbing up or down the A-frame on an obstacle
course? Difficulty may be defined as hesitation, repeatedly attempting to get down,
refusal to climb, crying out when performing the activity, or slipping or falling during
the climb.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
7. Does the dog show any difficulty performing 1-meter jumps on an obstacle course?
Difficulty may be defined as hesitation, refusal to jump, crying out when performing
the activity, not clearing the jump, or slipping or falling upon landing.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
50
8. Does the dog show any difficulty sitting from a standing position and/or standing
from a sitting position when performing this exercise 5 times in a row rapidly?
Difficulty may be defined as hesitation or reluctance to down/stand, refusal to down or
stand, or delayed or awkward changes in position when going into a down or rising
into a stand.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
9. Does the dog show any difficulty rising from a down position and/or going into
down from a standing position when performing this exercise 5 times in a row
rapidly? Difficulty may be defined as hesitation or reluctance to sit/stand, refusal to
sit or stand, or delayed or awkward changes in position when sitting or rising into a
stand.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
10. Does the dog show any difficulty performing a 2-meter tunnel crawl on an obstacle
course? Difficulty may be defined as hesitation in going into or coming out of the
tunnel, repeatedly unsuccessful attempts at getting down or rising, refusal to enter the
tunnel or get up from the crawl position, or crying out when performing the activity.
a. Never
b. Rarely
c. Sometimes
d. Frequently
e. Always
51
Questionnaire Scoring and Score Categorization
A score value of 0 to 4 will be assigned to each answer choice for Question 1 in the
following manner:
o Answers of “0 seconds” will receive a score of 4
o Answers of “1-2 seconds” will receive a score of 3
o Answers of “3-4 seconds” will receive a score of 2
o Answers of “5-6 seconds” will receive a score of 1
o Answers of “7 or more seconds” will receive a score of 0
A score value of 0 to 4 will be assigned to each answer choice for Questions 2-10 in
the following manner:
o “Never” receives a score of 0
o “Rarely” receives a score of 1
o “Sometimes” receives a score of 2
o “Frequently” receives a score of 3
o “Always” receives a score of 4
Scores from all questions from which a score from 0 to 4 was obtained will be
totaled.
The total questionnaire score for each Military Working Dog will be categorized as
follows:
o A total score less than or equal to 4 will be classified as “absence of functional
evidence for back pain.”
o A total score greater than or equal to 8 will be categorized as “presence of
functional evidence for back pain.”
o A total score of 5 to 7 will be considered as “borderline for presence of
functional evidence for back pain” and will not be considered sufficient for
dogs to be categorized in the LS pain group.
52
Appendix C: Exercise Protocol
The following describes the 8-week exercise protocol for the treatment group (LS Pain present
with exercise):
Exercises are to be performed twice per week on non-consecutive days. One exercise session
will precede each 8-week intervention program for all dogs consisting of brief (30-second to 1-
minute) trials with the various exercises for familiarization with the equipment, and
determination of the dog’s functionality for program individualization if necessary. These
exercises will be in addition to the dog’s normal walking routine as part of normal training aid
husbandry. Exercises should be performed continuously for each session. Each session is
expected to take approximately 35-45 personnel minutes total.
Weeks 1 and 2 Goals: Isometric Focus and Light Conditioning (Level I)
Exercise Routine:
a. Walk-trot intervals (one min each) on a land treadmill at zero incline for 15
minutes. Warm up for first three minutes and cool down for last 2 minutes.
b. Walking at a moderate pace through weave poles (5-6 poles) spaced at the same
length as the dog’s body, 10 times in each direction.
c. Stand/Sit/Down/Roll for 5 reps each on a mildly unstable surface, ie mattress.
Use a reward.
d. Standing on a large (85cm diameter) physioroll peanut for 5 minutes with
personnel stabilization at one end of the roll, providing very small movements of
the ball in a bouncing motion towards the floor, and rocking from side to side.
e. Step up and down exercises (have the dog step one forelimb at a time onto a
stairstep, then back down one limb at a time. Repeat up to 10 times as tolerated.
f. Standing on a flat surface with one forelimb and the opposite hindlimb raised
while maintaining the dog in a square position, preventing weight-shifting to
accommodate for the lifted limbs. Perform this exercise for 30 seconds at a time
with 30 second breaks between, for 10 repetitions. Easier variation: raise one
forelimb or hindlimb alone.
g. Standing on a flat surface and leaning in the following directions for a cookie, 10
repetitions:
i. Head to between front feet
ii. Head to left hind foot
iii. Head to right hind foot
a. Head straight up at 90 degrees from the dog’s longitudinal
axis
53
Weeks 3 and 4 Goals: Increasing Strength and Endurance Level I
Exercise Routine:
a. Walk-trot intervals (2 minutes each) on a land treadmill at zero incline for 15
minutes. Warm up for first three minutes and cool down for last 2 minutes.
b. Trotting through weave poles (6-8 poles) spaced at the length of the dog’s body,
10 times in each direction.
c. Stand/Sit/Down/Roll for 10 reps in each direction on a mildly unstable surface, ie
mattress pad. Use a reward.
d. Standing on a large (85cm diameter) physioroll peanut for 2 sets of 5 minutes
each with personnel stabilization at one end of the roll, providing very small
movements of the ball in a bouncing motion towards the floor, and rocking from
side to side.
e. “Hup” exercises: With dog’s forelimbs elevated on a steady surface at
approximately 4 feet to dog’s shoulder in height, have the dog follow a treat while
leaning right, left, forward and back without changing position of the hindlimbs.
Hold the treat for 3-5 seconds in each direction with intermittent rewarding. Do
this for 5 repetitions (allowing the dog to resume normal standing position off the
hup between reps) in each of the 4 directions.
f. Planks with forelimbs resting on a medium-sized peanut-shaped physioroll
(approximately 70cm diameter at widest point of the peanut). Hold for 30
seconds, with 30-second to 1-minute breaks between, for 10 repetitions.
g. Step up and down exercises (have the dog step one forelimb at a time onto a
stairstep, hold for 10 seconds, then back down one limb at a time. Repeat up to
10 times as tolerated.
h. “Superman” standing on a flat stable surface with contralateral forelimb and
hindlimb raised and gentle displacement forces in sagittal plane (cranial-caudal)
and frontal plane (axial-abaxial) applied. 30 seconds at a time with 30 second
breaks between, for 10 repetitions.
54
Weeks 5 and 6 Goals: Controlled concentric/eccentric, Dynamic Exercises and Moderate
Conditioning (Level II)
Exercise Routine:
a. Walk-trot intervals (1 minute each) on a land treadmill at a mild incline (5
degrees) for 15 minutes. Warm up for first three minutes and cool down for
last 2 minutes.
b. Trotting on leash through weave poles (6-8 poles) spaced at 2/3 of the length
of the dog’s body, 15 times in each direction.
c. Stand/sit/beg/down/roll for 2 sets of 5 reps (one set in each direction) on a
mattress. With each sit, have the dog go slowly into a hup/beg position and
hold for about 1 second before lowering back into the sit and standing again
for the next rep.
d. “Hup” exercises: With dog’s forelimbs elevated on a steady surface at
approximately 4 feet to dog’s shoulder in height, have the dog follow a treat
while leaning right, left, forward and back without changing position of the
hindlimbs. Hold the treat for 3-5 seconds in each direction with intermittent
rewarding. Do this for 10 repetitions (allowing the dog to resume normal
standing position off the hup between reps) in each of the 4 directions.
e. Standing on a large (85cm diameter) physioroll peanut for 5 minutes with
personnel stabilization at one end of the roll, providing small movements of
the ball in a bouncing motion towards the floor, and rocking from side to side.
f. Three-legged standing on the large physioroll peanut for 5 minutes, with one
forelimb raised for 2.5 minutes and one hindlimb raised for the remaining 2.5
minutes. This exercise will require 2 people.
g. Forelimbs on stable surface level with a Fit Pawz donut, hind limbs on donut,
balance for 5 minutes
h. Plank-to-push up exercises with 85cm diameter physioroll peanut: With the
dog’s hindlimbs planted on a mattress, have the dog’s forelimbs up on the
physioroll. Gently roll the physioroll cranially and caudally so the dog’s
antebrachii alternatively rest on the roll when in a more forward position, and
paws rest on the roll when in a more backward position. The ball movement
should be small so the hindlimbs stay in position. Perform this exercise for 2
sets of 5 reps in each direction.
55
Weeks 7 and 8 Goals: Increasing Strength and Endurance Level II
Exercise Routine:
a. Walk-trot intervals (2 minutes each) on a land treadmill at a mild incline (10
degrees) for 15 minutes. Warm up for first three minutes and cool down for
last 2 minutes.
b. Trotting on leash through weave poles (6-8 poles) spaced at the 2/3 of the
length of the dog’s body, 15 times in each direction.
c. Stand/sit/down/roll for 2 sets of 5 reps each on a mattress, ensuring every
other set is in the opposite direction. With each sit, have the dog go slowly
into a beg position and hold for about 2 seconds before lowering back into the
sit and standing again for the next rep.
d. Standing on a large (85cm diameter) physioroll peanut for 5 minutes with
personnel stabilization at one end of the roll, providing small movements of
the ball in a bouncing motion towards the floor, and rocking from side to side.
e. Three-legged standing on the large physioroll peanut for 5 minutes, with one
forelimb raised for 2.5 minutes and one hindlimb raised for the remaining 2.5
minutes. This exercise will require 2 people.
f. With hindlimbs on a donut and forelimbs on a stable surface at the same
height as the donut, balance for 5 minutes
g. Sitting/standing for 2 sets of 5 reps each with the dog entirely on a physioroll
peanut. Use a reward.
h. Plank-to-push up exercises with 85cm diameter physioroll peanut: With the
dog’s hindlimbs planted on a mattress, have the dog’s forelimbs up on the
physioroll. Gently roll the physioroll cranially and caudally so the dog’s
antebrachii alternatively rest on the roll when in a more forward position, and
paws rest on the roll when in a more backward position. The ball movement
should be small so the hindlimbs stay in position. Perform this exercise for 2
sets of 10 reps in each direction.
i. Step up and down exercises (have the dog step one forelimb at a time onto a
first and second stairstep, hold for 5 seconds, then back down one limb at a
time. Repeat 10 times as tolerated.
56
Vita
Dr. Andrea Henderson was born in Roanoke, VA and attended the Roanoke Valley Governor’s
School for Math and Science and Lord Botetourt High School, graduating in 1996. She
graduated with distinction with a Bachelor of Arts in Biology at the University of Virginia in
2000. Dr. Henderson was accepted into veterinary school and received her Doctorate of
Veterinary Medicine from the Virginia-Maryland Regional College of Veterinary Medicine in
2005. She participated in a Summer Fellowship Program in 2002 in which she conducted a
research project involving working dog and handler interactions. Her passion for working dogs
led her to join the U.S. Army Veterinary Corps in 2003, swearing in to active duty in May 2005.
She has been an active duty officer for nine years and currently holds the rank of Major. Her
first assignment was an internship at the Department of Defense Military Working Dog
Veterinary Services in San Antonio, Texas. She has since held leadership positions overseeing
the Mine Detector Dog program at Fort Leonard Wood, Missouri and the Kaiserslautern Branch
Veterinary Services in Germany. In July 2011, Dr. Henderson began a residency in canine sports
medicine and rehabilitation at the University of Tennessee in Knoxville, and has completed all
requirements of her residency in good standing. She is concurrently pursuing a Master of
Science with a focus in Kinesiology with plans to graduate in December 2014. In fall 2014 she
returns to the DoD Military Working Dog Center in San Antonio to oversee and expand the
sports medicine and conditioning programs for the military working dogs. Dr. Henderson has
specific interests in therapeutic laser applications in healing and nerve regeneration,
perioperative multimodal analgesia, and conservative management of degenerative lumbosacral
stenosis in Military Working Dogs.